1

I went outside last night and made some... pictures.

The details about my setup:

  • Sony alpha 6000
  • Varexon 2.8/135mm lens (would be 202.5mm on 36mm film)
  • ISO 25600
  • 3 seconds (a bit too long, given the Rule of 500)
  • stacking of about 10 images with lxnstack

Given that I was able to record 10.20 mag stars with the same lens and a single 2 s and ISO AUTO (= 5000 in that case) exposure - these faint stars appear as pale blue specks, I expected much better results with a higher ISO value. But no. They seem to be washed out.

Merak (beta Ursa Majoris), the Surfboard Galaxy and the Owl Nebula: enter image description here

I've seen tips to use the highest ISO when doing astrophotography, but it seems detrimental. What are some useful hints to find the right ISO value?

  • I'm assuming you've looked into long exposure instead of just cranking ISO? – Matthew Feb 13 '18 at 15:34
  • Yes, I use the longest exposures possible - while avoiding star trails... – user258532 Feb 13 '18 at 16:20
3

As you increase the ISO, your dynamic range decreases. While your camera has a base dynamic range around 13 stops, when you set the ISO to 25600 you only have a dynamic range of a little under 7. This means that bright stars will wash out much easier. This graph shows the effect of ISO on dynamic range for your camera. Sony a6000 dynamic range in relation to ISO Source: https://www.dxomark.com/Cameras/Sony/A6000---Measurements

To properly balance your ISO value you want to use the "500 rule". In a nut shell:

500 / [focal length] = maximum exposure time (to avoid star trails)

For your lens: 500/202 ≃ 2.5 seconds

Now that you know the shutter speed, open the aperture all the way and drop the ISO as low as you can.

Shooting at such a long focal length means that you might want to make a "barn door tracker" or buy a telescope tripod that adjust in live time. This will keep your camera pointing at the same place in relation to the sky, allowing you to put the ISO all the way down to 100 and shoot a much longer exposure without the introduction of star trails.

  • 1
    Ahh, I didn't really understand that high ISO will destroy dynamical range. I think that's why I was able to recover faint 10.2 mag stars in post-processing at ISO 5000, while the same was next to impossible at ISO 25600. And yes, making a barn door tracker is the next project on my list! – user258532 Feb 13 '18 at 16:29
  • Would I be right in thinking that the process of stacking and averaging frames will also reduce dynamic range? I'm thinking that a black pixel averaged with a number of noisy pixels will be more than true black and a white will be less than true white. To get black and white again the histogram will have to be stretched. The number of frames and the proportion of noisy pixels to good pixels in the set will alter the effect. – dmkonlinux Feb 13 '18 at 18:03
  • @dmkonlinux I can't see stacking and averaging reducing the dynamic range. The dynamic range will be reduced in that process, but not because of the stack. Instead it will be reduced because the image output won't be in full quality RAW. – Hairy Dresden Feb 13 '18 at 18:06
  • @hairy-dresden Sorry I pressed enter before finishing my train of thought. – dmkonlinux Feb 13 '18 at 18:08
  • @dmkonlinux Upon reading the rest of what you wrote, I agree with your logic. It wouldn't be by a very large amount, but it would certainly reduced the dynamic range a little. – Hairy Dresden Feb 13 '18 at 18:11
1

The on-die noise reduction done by some cameras before analog to digital conversion is what gives some cameras the moniker "star eater." This might be a case of the camera using more NR at ISO 25600 than at ISO 5000.

From this answer to What should I look for in a camera for shooting in bulb mode for astrophotography?:

Noise Reduction: Some cameras are known as "star eaters" because the way they do noise reduction on the sensor can eliminate dimmer stars as random noise. Nikon models with Hot Pixel Suppression (HPS) have been called 'star eaters' by some astrophotography enthusiasts. Ditto, for different reasons involving only bulb exposures, for most of Sony's recent A7 series of cameras. There's current debate going on, based on observations from pre-production testers, about whether the new yet-to-be released in the wild Sony A7 R III is a star eater or is not a star eater. For a long period many astro enthusiasts preferred Canon DSLRs because they did not do this type of NR on the sensor itself. Since the introduction of a new sensor in the EOS 80D in 2016, though, Canon appears to have moved in the same direction as what Sony and Nikon have been doing for quite a few years with regard to on-die noise reduction.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.